- Email: [email protected]
Ca exchange in the pancreatic B cell
Biochemical
Pharmacology, Vol. 45. No. 1, pp. 7-11. 1993. Printed in Great Britain.
DlO6-2952193$6.00 + 0.00 @I 1993. Pergamon Press Ltd
A ROLE FOR Na/Ca EXCHANGE IN THE PANCREATIC B CELL STUDIES WITH THAPSIGARGIN
AND CAFFEINE
ANDRE HERCHLJELZ*and PHILIPPE LEBRUN Laboratory of Pharmacology, Brussels University School of Medicine, 808 route de Lennik, Brussels, B-1070 Belgium (Receiued 19 June 1992; accepted 12 October 1992) Abstract-Sodium/calcium
(Na/Ca) exchange is thought to play a role in Ca*+ extrusion from the pancreatic B cell. The aim of the present study was to provide direct evidence for such a role. The effect of extracellular Na+ (Naof) removal on cytosolic free Ca*+ ([Ca”]J in single pancreatic B cells was examined using fura 2 and dual wavelength rnicrofluorimetry. Isosmotical replacement of Na$ by sucrose increased [Caz+li in the presence of extracellular Ca *+ but failed to affect [Ca2’li in the absence of the divalent cation. Thapsigargin (1 PM), an inhibitor of the endoplasmic reticulum Ca*+-ATPase, induced a transient increase in [Ca*+],in the presence of NaJ. This increase was enhanced and more sustained in the absence of Na$. In the absence of Na$ and the presence of thapsigargin, reintroduction of NaJ induced a rapid decrease in [Ca2+li.A similar picture was observed when caffeine (10 mM) was used to release Ca2+ from the endoplasmic reticulum. The decrease in [Ca2+li induced by Na$ reintroduction was accompanied by an important increase in 4SCa outflow from perifused islets. In conclusion, this study provides direct evidence that Na/Ca exchange may regulate B cell [Ca2’li within physiological range.
In several types of cells, Na/Ca exchange is thought to represent an important mechanism for Ca2+ extrusion to the extracellular milieu [l-3]. Because in excitable cells the process is electrogenic and sensitive to membrane potential [45], it may reverse and also drive Ca2+ inflow during electrical activity [3]. In the pancreatic B cell, the existence of a Na/ Ca exchange has been postulated for many years [68]. Moreover, based on indirect evidences, it has been proposed that the process of Na/Ca exchange could participate in Ca2+ extrusion from the B cell [6-91. However, direct evidence for such a role is lacking. The aim of the present study was to provide direct evidence for a modulatory role of Na/Ca exchange on [Ca2’]i in the pancreatic B cell. For such a purpose, the effect of extracellular Na+ removal on [Ca +]i in single rat pancreatic B cells was examined using fura 2 and dual wavelength microfluorimetry.
incubated with Fura 2 AM (final concentration: 4 yM) for 1 hr and, after washing, the coverslips with the cells were mounted as the bottom of an open chamber placed on the stage of the microscope. Fura 2 fluorescence of single loaded cells was measured using dual excitation microfluorimetry with a Spex photometric system (Optilas, Alphen aan den Rjn, Holland). The system was coupled to an inverted fluorescence microscope (Diaphot TDM, Nikon, Tokyo, Japan) equipped with Fluor objectives (CF 40x and CF 100x, Nikon) for epi-fluorescence. The excitation wavelengths (340 and 380nm) were alternated at a frequency of 1 Hz, the length of time for data collection at each wavelength being 0.05 sec. The emission wavelength was 510nm. [Ca2+]i was calculated from the ratios of the 340 and 380nm signals after background subtraction using the equation:
tR -
MATERIALSANDMETHODS Fluorescence measurement of single cells. The method used to isolate islet cells from rat pancreas has been described elsewhere [lo]. The preparation has a cell viability of about 97% and responds satisfactorily to various insulin secretagogues [ 10,111. The cells were placed on glass coverslips and maintained in tissue culture for 48-72 hr before use, as described previously [12]. The cells were then * Corresponding author: A. Herchuelz, Laboratoire de Pharmacodynamie et de Therapeutique, Universitt Libre de Bruxelles, Facultd de Mbdecine, BLtiment GE, 808 route de Lennik, B-1070 Bruxelles. Tel. (32) 2 555 62 01; FAX (32) 2 555 63 70.
&tin)
Sf2
b2+li = Kf x cRmax _ Rj XQ
where Kd is the dissociation constant for the Fura 2-Ca2+ complex (224nM at 37”), R is the ratio of fluorescence at 340 over 380nm, Sf2 is the fluorescence of free dye measured at 380 nm and S&! is the fluorescence of bound dye measured at 380 nm. RIlla,R,i, and Sf2/&,2 were determined in separate experiments by recording Fura 2 fluorescence in the absence of extracellular Caz’ or in the presence of a saturating Ca2+ concentration [12]. The open chamber (1 mL) was thermostated at 37” and perfused at a rate of 2mL/min. In most cases, the stimuli were applied to the cells using the perfusing system. In the case of drugs available in very small amounts (thapsigargin), the compound was injected
8
directly into the chamber as a small aliquot and the perfusion stopped. The drug was removed from the chamber by switching on the perfusion system. The period of equilibration before starting fluorescence measurements was 15 min. 45Ca outpow from perifused islets. The method used to measure 45Ca outflow from perifused islets has been described previously [7]. Briefly, groups of 100 islets were incubated for 60 min in the presence of 16.7mM glucose and 45Ca (0.02-O.O4mM, lOO&i/mL). After incubation, the islets were washed three times and then placed in a perifusion chamber. The efflux of 45Ca (cpm/min) was expressed as a fractional outflow rate (% lost per min). The medium used to perfuse the cells and the islets was a Krebs-Ringer bicarbonate buffered solution having the following composition (in mM):NaCl 115, KC1 5, CaCl* 2.56, MgCl* 1, NaHC03 24. The solution was equilibrated against a mixture of O2 (95%) and CO2 (5%) and for the isolated cells also contained HEPES-NaOH (10 mM, pH 7.4). Some media contained no Ca*+ and were enriched with 0.5 mM EGTA. In some experiments, NaCl was isosmotically replaced by sucrose (200 mM) and NaHC03 by choline bicarbonate. The latter media also contained atropine 3.6pM [7] to avoid cholinergic effects. In some other experiments, NaCl and NaHC03 were isosomotically replaced by sucrose (241 mM) or by choline chloride (139 mM) and HEPES-NaOH (10 mM) was replaced by HEPES-KOH (10 mM). The media also contained, when required, thapsigargin (Gibco BRL, Gent, Belgium) or caffeine (UCB , Leuven, Belgium). The media contained no glucose except for the experiments carried out in the presence of caffeine that contained 2.8 mM glucose. The statistical significance of differences between mean data was assessed by using the Student’s t-test. The area under the concentration-time curve (AUC) was measured using the trapezoidal rule. RE!WLTSANDDISCUSSION
As a first step, we have examined the effect of extracellular Na+ (Na$) removal on [Ca*‘]i of B cells perfused in the presence of extracellular Ca*+. Isosmotical replacement of Na$ by sucrose increased [Ca*+]i (Fig. l), the increase being variable from cell to cell. In a series of 45 cells, 24 (53%) displayed a biphasic increase in [Ca*‘]i (Fig. la), consisting of an initial phase followed by a plateau phase. The initial phase lasted 30-60sec. Nine cells (20%) displayed a square wave increase in [Ca*‘]i (Fig. lb), while 12 (27%) displayed a modest increase in [Ca”]i onto which distinct spikes were superimposed (Fig. lc). In each case, the effect of Nao+ removal was rapidly reversible. Similar effects of Nao+removal were observed whether in the presence or absence of bicarbonate, or whether Na+ was replaced by sucrose or choline. When the same experiments were carried out in the absence of extracellular Ca*+, Na$ removal failed to affect [Ca*+]i (Fig. Id). These data provide direct evidence that activation of reverse Na/Ca exchan e by changing the Na+ gradient may increase [Ca f +]i in the B cell. Indeed, Nao+removal has been shown to increase 45Ca uptake
in islet cells by reverse Na/Ca exchange [ll]. The Ca*’ response was heterogeneous like that to many other secretagogues [12]. Incidentally, it cannot be excluded that the heterogeneity of the responses reflects in part the responses of different types of islet cells instead of B cells, although much care was taken to select only B cells according to their size [12]. The Ca*+ responses were also biphasic in the majority of cases. In Ca*+ uptake experiments, an initial fast component lasting 30-60 set followed by a slower increase in Ca*+ inflow is also observed when NaJ is removed [ll]. That the increase in [Ca*+]i resulted from a rise in Ca*+ inflow (by reverse Na/Ca exchange) was confirmed by the disappearance of the increase in the absence of extracellular Ca*+. The latter observation also suggests that at a low [Ca*‘]i, as induced by a prolonged ex osure (+20 min) to a Ca*+-depleted medium, Na PCa exchange does not play a major role in Ca*+ extrusion from the B cell. Thus, if Na/ Ca exchange played a significant role in the control of [Ca*+]i below basal level (about lOOnM), the removal of NaJ in the absence of Ca*+,, would be expected to increase [Ca*+]i by inhibiting (forward) Na/Ca exchange. In the absence of Ca*+,,, basal [Ca”]i averaged 65 2 11 nM (N = 22) compared to 120 -C 10 nM (N = 58) in the presence of extracellular Ca*+. This observation is in agreement with the low affinity characteristics of the Na/Ca exchanger for Ca*+ in many type of cells [ 1,131. In the next series of experiments, we have examined to what extent the presence of Na$ could help to reduce [Ca*‘]i. This was examined under raised [Ca*+]i. TO raise [Ca*+]i, we used two drugs known to interfere with the Ca*+ uptake and release mechanisms in the endoplasmic reticulum: thapsigargin and caffeine. Thapsigargin in a specific blocker of the Ca*+-ATPase of the endoplasmic reticulum [14]. Caffeine sensitizes the Ca*+-induced Ca*+ release channels of the endoplasmic reticulum to resting Ca *+ levels [15]. In the presence of Na+,, (139 mM), thapsigargin (1 PM) induced a rapid but short-lived increase in [Ca*+]i, the increase being maximal during the first 3 min of exposure to thapsigargin (Fig. 2, upper panel). When the B cells were perfused in the absence of Naof from 20 min, basal [Ca*+]i was not elevated compared to controls (Fig. 2, lower panel, P > 0.05). This was not unexpected since with increasing time of exposure to low Na+ medium, a fall in the intracellular Na+ concentration occurs that reduces the Na+ gradient and hence Ca*+ entry into the cell by reverse Na/ Ca exchange [ 111. In the absence of Nao+, thapsigargin also increased [Ca*+]i but the effect was more sustained (Fig. 2, lower panel) than in the presence of Na$. Indeed, the increase in [Ca*+]i, as evaluated by measurement of the area under the curve, while not being affected during the first 3 min of exposure to thapsigargin (P > 0.01) was increased by 154 -t 46% (N = 7) and 127 * 35% (N = 5) during the first 6 or 9min, respectively (P < 0.02 and P < 0.04). In further experiments, we examined the effect of a change in [Na+10 on [Ca*+]i of a cell exposed to thapsigargin. Figure 3 (lower panel) shows that normalizing the [Na+],, induced a rapid drop of
Na/Ca exchange in the pancreatic B cell I
[CPI
csz+1
aft?’ 30
9
A
*‘B
Na' 0
300
“I
1
Na+0
Nat0
TIME bin)
Fig. 1. Effect ofN4 removal on [Ca2+ji in single B cells. The cells were perfused in the presence (A, B, C) or the absence of extracellular Ca2” (D). The panels A-C represent different typical patterns of [Caz+Ji changes. The bar indicate the period of exposure to the absence of Na;t. The traces are representative of 24 (A), 9 (B), 12 (C) and 14 (D) cells.
ICa271 nM
I Ca*+li
nM
THAPSlGARGlN
THAPSIGAAGIN
1Y M
Na*O 5oo Na+ 0 THAPSI
Na’O
,.
1Y M
Fig. 2. Effect of tha~igar~n (1 FM) on (Caz+]i in single B cells. The cells were perfused either in the presence (upper panel) or the absence (lower panel) of NaJ. The bar indicates the period of exposure to thapsigargin. The traces are representative of seven (upper panel) and five (lower panel) individual experiments.
ttf& &a&t) Fig. 3. Effect of thapsigargtin (thapsi) on [Ca’“ji in sin&e B cehs perfused in the absence of NaO+.In the Iower panel, Na+ (139 mM) was reintroduced at the time indicated. The traces are representative of 16 (upper panel) and 21 (lower panel) individual experiments.
A. HERCHUELZ and P. LEBRUN
10
Fig. 4. Effect of caffeine (1OmM) on [Caz+]jin single B cells perfused in the absence of Nao+. Na+ (139mM) was reintroduced at the time indicated. Basal media contained 2.8 mM glucose. The trace is representative of 19 individual experiments.
025t
.
30
) , ( , , ) 40 50 60 70 80 90 Time &tin)
very fast and short-lived spikes on an elevated Cat+ level (data not shown). In the absence of Na$, caffeine most frequently caused a biphasic increase in [Ca’+]i with a short initial phase followed by a second plateau phase (Fig. 4). Reintroduction of NaJ during this phase produced an immediate drop in [Ca’+]i. It is interesting to notice that after Na$ reintroduction, caffeine still had a stimulatory effect on [Ca*‘]i in the form of sinusoidal oscillations on a slightly elevated [Ca2+ii level (Fig. 4). To ascertain that the changes in [Ca2”]i seen on Na+ reintroduction were indeed due to changes in Ca2+ outflow, we examined *Va outflow from perifused islets exposed to caffeine. In the presence of both extracellular Ca2+ and Na+, caffeine induced a rapid increase in 45Ca outllow from perifused islets (Fig. 5, left panel). This increase was not suppressed in the absence of extracellular Ca*+ (Fig. 5, left panel), suggesting that it resulted from the release
0.25t , I ( , . , , 30 40 50 68 70 80 90 Time (min)
Fig. 5. Effect of caffeine on “Ca outflow from perifused islets. Left panel: The islets were perifused either in the presence (0) or the absence (0) of CaZfo (2.56 mM). Right panel: The islets were perifused in the presence of Ca 2+~but absence of Na$. The closed circles and solid lines depict experiments in which NaCl was reintroduced in the extracellular milieu at the 70th min of perifusion. Basal media contained 2.8 mM glucose.
[Ca2’fr in a cell perfused in the absence of Na$ from 20 min and thapsigargin from about 3 min. In control experiments, where the absence of [Naf10 was maintained throughout, such a rapid drop was not observed (Fig. 3, upper panel). The regression line characterizing the decrease in [Ca’+]r before (1.5 min) and after (3 min) thapsigargin removal was examined in two series of cells displaying comparable basal [Ca”]i and peak increases in fCa2+li in response to thaps~gar~n (P > 0.1 and 0.6, resistively). When Na+ was reintroduced at the time of thapsigargin removal, a significant change in the slope was observed (P < 0.001). In contrast, no significant change in the slope was observed when the absence of extracellular Na+ was maintained at the time of thapsigargin removal (P > 0.4). As compared to that of thapsigargin, the effect of caffeine on [Ca2+li was more heterogeneous. It was also characterized by the frequent occurrence of
of Ca” from intracellular stores [15]. By contrast, in the presence of Ca*+ and the absence of Nao+(Fig. 5, right panel), the increase in “%a outflow induced by caffeine was almost completely suppressed. The figure shows that reintroduction of Na+ into the extracellular milieu induced a marked increase in *%Zaoutflow from islets exposed to caffeine as in the case of thapsigargin. These data indicate that in the presence of extracelluar Na’, the Ca2+ released by caffeine from in~acelIular stores was extruded from the ceil presumably by Na/Ca exchange. When such a process was blocked by the absence of extracellular Na+, higher [Cs2”]i levels were reached in response to caffeine as in the case of thapsigargin. When the process was reactivated by Nao+ readmission, 45Ca outflow was stimulated (Fig. 5 right panel) and [Ca*‘]i dropped (Fig. 4). The removal of extracellular Na+ may not solely inhibit Na/Ca exchange but could also interfere with
11
Na/Ca exchange in the pancreatic B cell
other cellular processes, e.g. Na/H exchange, Na+,K+-ATPase, membrane potential. . . However, alterations in the latter processes do not appear to be responsible for the observed changes in [Ca*+]i and 45Ca outflow Indeed, at low glucose concentrations (~2.8 mM) , extracellular Na+ removal only barely affects membrane potential [16]. N& removal could inhibit Na+,K+-ATPase, but this would not lead to Na+i accumulation and hence would not affect cellular Ca2+ movements [7, 171. Na$ removal could also inhibit Na/H exchange and result in a decrease in pHi [18]. This acidification may reduce Ca2+ outflow but to a much lower extent than Nat removal [19]. In addition, the decrease in pHi reduces 45Ca outflow by inhibiting Na/Ca exchange [ 191. Our data provide direct evidence that Na/Ca exchange, working in its Ca2+ efflux mode, may regulate the [Ca2+]i of the pancreatic B cell. Thus, the presence of extracellular Na+ favoured the decrease in [Ca*+]i that had been raised by two different mechanisms. Our data strongly indicate that the presence of Nao+probably acted by increasing Ca2+ outflow from the cell. This conclusion is strengthened by the observation that the absence of Na$ affected the late instead of the immediate effects of thapsigargin on [Ca2+]i. In other words, the absence of NaJ did not affect the action of the drug (blockade of Ca2+ uptake by the endoplasmic reticulum). Rather, it interfered with a cellular homoeostatic mechanism allowing the B cell to recover from a cytosolic Ca*+ load. Our data are in agreement with recent patchclamp experiments combined with microfluorimetric recordings in mouse pancreatic B cells showing that Na+ removal produced both an elevation of resting [Ca2+]i and an increased amplitude of depolarization-evoked [Ca2+]i transients [20]. In conclusion, the B cell Na/Ca exchange appears as a Ca pumping system designed to extrude Ca*+ when its intracellular concentration is raised above basal levels, as in many other types of cell. Further work is required to determine the exact kinetic constants of the exchange and the possible participation of the process in Ca2+ uptake mechanisms. Acknowledgements-The
authors are indebted to Ch. Pastiels, M. Hermann and A. Van Praet for technical help, and to P. Surardt for secretarial help. REFERENCES 1. Dipolo Rand Beau& L, Ca2+ transport in nerve fibers. Biochim Biophys Acta 947: 549-569, 1988. 2. Reeves JP and Philipson KD, Sodium-calciumexchange activity in plasma membranes vesicles. In: Sodiumcalcium Exchange (Eds. Allen TJA, Noble D and
Reuter H), pp. 27-53. Oxford Science Publications, Oxford, 1989
3. Sheu SS and Blaustein MP, Sodium/calcium exchange and regulation of cell calcium and contractility in cardiac muscle, with a note about vascular smooth muscle. In: The Heart and the Cardiovascular System (Ed. Fozzard HA), pp. 50!&535. Raven Press, New York, 1986. 4. Eisner DA and Lederer WJ, Na-Ca exchange: stoichiometry and electrogenicity. Am J Physiol 248: C189-C202, 1985. 5. Mullins LJ, The generation
of electric currents in cardiac fibers by Na-Ca exchange. Am J Physiol236: c103-CllO, 1979. 6. Hellman B, Andersson T, Berggren PO and Rorsman P. Calcium and nancreatic &cell function XI. Modification of “Ca’fluxes by Na removal. Biochem Med 24: 143-152, 1980. 7. Herchuelz A, Sener A and Malaisse WJ, Regulation
of calcium fluxes in rat pancreatic islets: calcium extrusion by sodium-calcium countertransport. J Membr Biol57: 1-12, 1980. 8. Janjic D and Wollheim CB, Interactions of Ca*+, M$+,
and Na+ in regulation of insulin release from rat islets. E222-E229, 1983. 9. Siegel EG, Wollheim CB, Renold AE and Sharp GWG, Evidence for the involvement of Na-Ca exchange in glucose-induced insulin release from rat pancreatic islets. J Clin Invest 66: 9961003, 1980. 10. Gobbe P and Herchuelz A, Does glucose decrease cytosolic free calcium in normal pancreatic islet cells? Am J Physiol244:
Res Commun 1989.
Chem Path01 Pharmacol
63: 231-247,
11. Plasman PO, Lebrun P and Herchuelz A, Characterization of the process of sodium-calcium exchange in nancreatic islet cells. Am J Phvsiol259: E844-E850. , 1990. 12. Herchuelz A, Pochet R, Pastiels C and Van Praet A, Heterogeneous changes in [Ca2+]iinduced by glucose, tolbutamide and K+ in single rat pancreatic B cells. Cell Calcium 12: 577-586, 1991. 13. Carafoh E, Membrane transport of calcium: an overview. Methodr Enzymol157: 2-11, 1988. 14. Thastrup 0, Cullen PJ, Drobak PK, Hartley MR and
Dawson AP, Thapsigargin, a tumor promoter, discharges intracellular Ca2’ stores by specific inhibition of the endoplasmic reticulum Ca2+-ATPase. Proc Nat1 Acad Sci USA 87: 2466-2470,
1990.
15. Endo M, Calcium release from sarcoplasmic reticulum. Cm-r Top Membr Transp 25: 181-230, 1985. 16. Ribalet B and Beigelman PM, Effects of sodium on /Jcell electrical activity. Am J Physiol242: C296C303, 1982.
17. Lebrun P, Malaisse WJ and Herchuelz A, Na+-K+ pump activity and the glucose-stimulated Ca2+-sensitive K+ permeability in the pancreatic B-cell. J Membr Biol 74: 67-73, 1983.
18. Lebrun P, Plasman PO and Herchuelz A, Effect of extracellular sodium removal upon %Rb outflow from pancreatic islet cells. Biochim Biophys Acta 1011: 611, 1989. 19. Lebrun P, Malaisse WJ and Herchuelz A, Effect of the absence of bicarbonate won intracellular oH and calcium fluxes in pancreatic Islet cells. Biochim ‘Biophys Acta 721: 357-365,
1982.
20. Rorsman P, Ammall C, Berggren P-O, Bokvist K and Larsson 0, Cytoplasmic calcium transients due to single action potentials and voltage-clamp depolarizations in mouse pancreatic B-cells. Embo J 11: 2877-2884,1992.
Pharmacology, Vol. 45. No. 1, pp. 7-11. 1993. Printed in Great Britain.
DlO6-2952193$6.00 + 0.00 @I 1993. Pergamon Press Ltd
A ROLE FOR Na/Ca EXCHANGE IN THE PANCREATIC B CELL STUDIES WITH THAPSIGARGIN
AND CAFFEINE
ANDRE HERCHLJELZ*and PHILIPPE LEBRUN Laboratory of Pharmacology, Brussels University School of Medicine, 808 route de Lennik, Brussels, B-1070 Belgium (Receiued 19 June 1992; accepted 12 October 1992) Abstract-Sodium/calcium
(Na/Ca) exchange is thought to play a role in Ca*+ extrusion from the pancreatic B cell. The aim of the present study was to provide direct evidence for such a role. The effect of extracellular Na+ (Naof) removal on cytosolic free Ca*+ ([Ca”]J in single pancreatic B cells was examined using fura 2 and dual wavelength rnicrofluorimetry. Isosmotical replacement of Na$ by sucrose increased [Caz+li in the presence of extracellular Ca *+ but failed to affect [Ca2’li in the absence of the divalent cation. Thapsigargin (1 PM), an inhibitor of the endoplasmic reticulum Ca*+-ATPase, induced a transient increase in [Ca*+],in the presence of NaJ. This increase was enhanced and more sustained in the absence of Na$. In the absence of Na$ and the presence of thapsigargin, reintroduction of NaJ induced a rapid decrease in [Ca2+li.A similar picture was observed when caffeine (10 mM) was used to release Ca2+ from the endoplasmic reticulum. The decrease in [Ca2+li induced by Na$ reintroduction was accompanied by an important increase in 4SCa outflow from perifused islets. In conclusion, this study provides direct evidence that Na/Ca exchange may regulate B cell [Ca2’li within physiological range.
In several types of cells, Na/Ca exchange is thought to represent an important mechanism for Ca2+ extrusion to the extracellular milieu [l-3]. Because in excitable cells the process is electrogenic and sensitive to membrane potential [45], it may reverse and also drive Ca2+ inflow during electrical activity [3]. In the pancreatic B cell, the existence of a Na/ Ca exchange has been postulated for many years [68]. Moreover, based on indirect evidences, it has been proposed that the process of Na/Ca exchange could participate in Ca2+ extrusion from the B cell [6-91. However, direct evidence for such a role is lacking. The aim of the present study was to provide direct evidence for a modulatory role of Na/Ca exchange on [Ca2’]i in the pancreatic B cell. For such a purpose, the effect of extracellular Na+ removal on [Ca +]i in single rat pancreatic B cells was examined using fura 2 and dual wavelength microfluorimetry.
incubated with Fura 2 AM (final concentration: 4 yM) for 1 hr and, after washing, the coverslips with the cells were mounted as the bottom of an open chamber placed on the stage of the microscope. Fura 2 fluorescence of single loaded cells was measured using dual excitation microfluorimetry with a Spex photometric system (Optilas, Alphen aan den Rjn, Holland). The system was coupled to an inverted fluorescence microscope (Diaphot TDM, Nikon, Tokyo, Japan) equipped with Fluor objectives (CF 40x and CF 100x, Nikon) for epi-fluorescence. The excitation wavelengths (340 and 380nm) were alternated at a frequency of 1 Hz, the length of time for data collection at each wavelength being 0.05 sec. The emission wavelength was 510nm. [Ca2+]i was calculated from the ratios of the 340 and 380nm signals after background subtraction using the equation:
tR -
MATERIALSANDMETHODS Fluorescence measurement of single cells. The method used to isolate islet cells from rat pancreas has been described elsewhere [lo]. The preparation has a cell viability of about 97% and responds satisfactorily to various insulin secretagogues [ 10,111. The cells were placed on glass coverslips and maintained in tissue culture for 48-72 hr before use, as described previously [12]. The cells were then * Corresponding author: A. Herchuelz, Laboratoire de Pharmacodynamie et de Therapeutique, Universitt Libre de Bruxelles, Facultd de Mbdecine, BLtiment GE, 808 route de Lennik, B-1070 Bruxelles. Tel. (32) 2 555 62 01; FAX (32) 2 555 63 70.
&tin)
Sf2
b2+li = Kf x cRmax _ Rj XQ
where Kd is the dissociation constant for the Fura 2-Ca2+ complex (224nM at 37”), R is the ratio of fluorescence at 340 over 380nm, Sf2 is the fluorescence of free dye measured at 380 nm and S&! is the fluorescence of bound dye measured at 380 nm. RIlla,R,i, and Sf2/&,2 were determined in separate experiments by recording Fura 2 fluorescence in the absence of extracellular Caz’ or in the presence of a saturating Ca2+ concentration [12]. The open chamber (1 mL) was thermostated at 37” and perfused at a rate of 2mL/min. In most cases, the stimuli were applied to the cells using the perfusing system. In the case of drugs available in very small amounts (thapsigargin), the compound was injected
8
directly into the chamber as a small aliquot and the perfusion stopped. The drug was removed from the chamber by switching on the perfusion system. The period of equilibration before starting fluorescence measurements was 15 min. 45Ca outpow from perifused islets. The method used to measure 45Ca outflow from perifused islets has been described previously [7]. Briefly, groups of 100 islets were incubated for 60 min in the presence of 16.7mM glucose and 45Ca (0.02-O.O4mM, lOO&i/mL). After incubation, the islets were washed three times and then placed in a perifusion chamber. The efflux of 45Ca (cpm/min) was expressed as a fractional outflow rate (% lost per min). The medium used to perfuse the cells and the islets was a Krebs-Ringer bicarbonate buffered solution having the following composition (in mM):NaCl 115, KC1 5, CaCl* 2.56, MgCl* 1, NaHC03 24. The solution was equilibrated against a mixture of O2 (95%) and CO2 (5%) and for the isolated cells also contained HEPES-NaOH (10 mM, pH 7.4). Some media contained no Ca*+ and were enriched with 0.5 mM EGTA. In some experiments, NaCl was isosmotically replaced by sucrose (200 mM) and NaHC03 by choline bicarbonate. The latter media also contained atropine 3.6pM [7] to avoid cholinergic effects. In some other experiments, NaCl and NaHC03 were isosomotically replaced by sucrose (241 mM) or by choline chloride (139 mM) and HEPES-NaOH (10 mM) was replaced by HEPES-KOH (10 mM). The media also contained, when required, thapsigargin (Gibco BRL, Gent, Belgium) or caffeine (UCB , Leuven, Belgium). The media contained no glucose except for the experiments carried out in the presence of caffeine that contained 2.8 mM glucose. The statistical significance of differences between mean data was assessed by using the Student’s t-test. The area under the concentration-time curve (AUC) was measured using the trapezoidal rule. RE!WLTSANDDISCUSSION
As a first step, we have examined the effect of extracellular Na+ (Na$) removal on [Ca*‘]i of B cells perfused in the presence of extracellular Ca*+. Isosmotical replacement of Na$ by sucrose increased [Ca*+]i (Fig. l), the increase being variable from cell to cell. In a series of 45 cells, 24 (53%) displayed a biphasic increase in [Ca*‘]i (Fig. la), consisting of an initial phase followed by a plateau phase. The initial phase lasted 30-60sec. Nine cells (20%) displayed a square wave increase in [Ca*‘]i (Fig. lb), while 12 (27%) displayed a modest increase in [Ca”]i onto which distinct spikes were superimposed (Fig. lc). In each case, the effect of Nao+ removal was rapidly reversible. Similar effects of Nao+removal were observed whether in the presence or absence of bicarbonate, or whether Na+ was replaced by sucrose or choline. When the same experiments were carried out in the absence of extracellular Ca*+, Na$ removal failed to affect [Ca*+]i (Fig. Id). These data provide direct evidence that activation of reverse Na/Ca exchan e by changing the Na+ gradient may increase [Ca f +]i in the B cell. Indeed, Nao+removal has been shown to increase 45Ca uptake
in islet cells by reverse Na/Ca exchange [ll]. The Ca*’ response was heterogeneous like that to many other secretagogues [12]. Incidentally, it cannot be excluded that the heterogeneity of the responses reflects in part the responses of different types of islet cells instead of B cells, although much care was taken to select only B cells according to their size [12]. The Ca*+ responses were also biphasic in the majority of cases. In Ca*+ uptake experiments, an initial fast component lasting 30-60 set followed by a slower increase in Ca*+ inflow is also observed when NaJ is removed [ll]. That the increase in [Ca*+]i resulted from a rise in Ca*+ inflow (by reverse Na/Ca exchange) was confirmed by the disappearance of the increase in the absence of extracellular Ca*+. The latter observation also suggests that at a low [Ca*‘]i, as induced by a prolonged ex osure (+20 min) to a Ca*+-depleted medium, Na PCa exchange does not play a major role in Ca*+ extrusion from the B cell. Thus, if Na/ Ca exchange played a significant role in the control of [Ca*+]i below basal level (about lOOnM), the removal of NaJ in the absence of Ca*+,, would be expected to increase [Ca*+]i by inhibiting (forward) Na/Ca exchange. In the absence of Ca*+,,, basal [Ca”]i averaged 65 2 11 nM (N = 22) compared to 120 -C 10 nM (N = 58) in the presence of extracellular Ca*+. This observation is in agreement with the low affinity characteristics of the Na/Ca exchanger for Ca*+ in many type of cells [ 1,131. In the next series of experiments, we have examined to what extent the presence of Na$ could help to reduce [Ca*‘]i. This was examined under raised [Ca*+]i. TO raise [Ca*+]i, we used two drugs known to interfere with the Ca*+ uptake and release mechanisms in the endoplasmic reticulum: thapsigargin and caffeine. Thapsigargin in a specific blocker of the Ca*+-ATPase of the endoplasmic reticulum [14]. Caffeine sensitizes the Ca*+-induced Ca*+ release channels of the endoplasmic reticulum to resting Ca *+ levels [15]. In the presence of Na+,, (139 mM), thapsigargin (1 PM) induced a rapid but short-lived increase in [Ca*+]i, the increase being maximal during the first 3 min of exposure to thapsigargin (Fig. 2, upper panel). When the B cells were perfused in the absence of Naof from 20 min, basal [Ca*+]i was not elevated compared to controls (Fig. 2, lower panel, P > 0.05). This was not unexpected since with increasing time of exposure to low Na+ medium, a fall in the intracellular Na+ concentration occurs that reduces the Na+ gradient and hence Ca*+ entry into the cell by reverse Na/ Ca exchange [ 111. In the absence of Nao+, thapsigargin also increased [Ca*+]i but the effect was more sustained (Fig. 2, lower panel) than in the presence of Na$. Indeed, the increase in [Ca*+]i, as evaluated by measurement of the area under the curve, while not being affected during the first 3 min of exposure to thapsigargin (P > 0.01) was increased by 154 -t 46% (N = 7) and 127 * 35% (N = 5) during the first 6 or 9min, respectively (P < 0.02 and P < 0.04). In further experiments, we examined the effect of a change in [Na+10 on [Ca*+]i of a cell exposed to thapsigargin. Figure 3 (lower panel) shows that normalizing the [Na+],, induced a rapid drop of
Na/Ca exchange in the pancreatic B cell I
[CPI
csz+1
aft?’ 30
9
A
*‘B
Na' 0
300
“I
1
Na+0
Nat0
TIME bin)
Fig. 1. Effect ofN4 removal on [Ca2+ji in single B cells. The cells were perfused in the presence (A, B, C) or the absence of extracellular Ca2” (D). The panels A-C represent different typical patterns of [Caz+Ji changes. The bar indicate the period of exposure to the absence of Na;t. The traces are representative of 24 (A), 9 (B), 12 (C) and 14 (D) cells.
ICa271 nM
I Ca*+li
nM
THAPSlGARGlN
THAPSIGAAGIN
1Y M
Na*O 5oo Na+ 0 THAPSI
Na’O
,.
1Y M
Fig. 2. Effect of tha~igar~n (1 FM) on (Caz+]i in single B cells. The cells were perfused either in the presence (upper panel) or the absence (lower panel) of NaJ. The bar indicates the period of exposure to thapsigargin. The traces are representative of seven (upper panel) and five (lower panel) individual experiments.
ttf& &a&t) Fig. 3. Effect of thapsigargtin (thapsi) on [Ca’“ji in sin&e B cehs perfused in the absence of NaO+.In the Iower panel, Na+ (139 mM) was reintroduced at the time indicated. The traces are representative of 16 (upper panel) and 21 (lower panel) individual experiments.
A. HERCHUELZ and P. LEBRUN
10
Fig. 4. Effect of caffeine (1OmM) on [Caz+]jin single B cells perfused in the absence of Nao+. Na+ (139mM) was reintroduced at the time indicated. Basal media contained 2.8 mM glucose. The trace is representative of 19 individual experiments.
025t
.
30
) , ( , , ) 40 50 60 70 80 90 Time &tin)
very fast and short-lived spikes on an elevated Cat+ level (data not shown). In the absence of Na$, caffeine most frequently caused a biphasic increase in [Ca’+]i with a short initial phase followed by a second plateau phase (Fig. 4). Reintroduction of NaJ during this phase produced an immediate drop in [Ca’+]i. It is interesting to notice that after Na$ reintroduction, caffeine still had a stimulatory effect on [Ca*‘]i in the form of sinusoidal oscillations on a slightly elevated [Ca2+ii level (Fig. 4). To ascertain that the changes in [Ca2”]i seen on Na+ reintroduction were indeed due to changes in Ca2+ outflow, we examined *Va outflow from perifused islets exposed to caffeine. In the presence of both extracellular Ca2+ and Na+, caffeine induced a rapid increase in 45Ca outllow from perifused islets (Fig. 5, left panel). This increase was not suppressed in the absence of extracellular Ca*+ (Fig. 5, left panel), suggesting that it resulted from the release
0.25t , I ( , . , , 30 40 50 68 70 80 90 Time (min)
Fig. 5. Effect of caffeine on “Ca outflow from perifused islets. Left panel: The islets were perifused either in the presence (0) or the absence (0) of CaZfo (2.56 mM). Right panel: The islets were perifused in the presence of Ca 2+~but absence of Na$. The closed circles and solid lines depict experiments in which NaCl was reintroduced in the extracellular milieu at the 70th min of perifusion. Basal media contained 2.8 mM glucose.
[Ca2’fr in a cell perfused in the absence of Na$ from 20 min and thapsigargin from about 3 min. In control experiments, where the absence of [Naf10 was maintained throughout, such a rapid drop was not observed (Fig. 3, upper panel). The regression line characterizing the decrease in [Ca’+]r before (1.5 min) and after (3 min) thapsigargin removal was examined in two series of cells displaying comparable basal [Ca”]i and peak increases in fCa2+li in response to thaps~gar~n (P > 0.1 and 0.6, resistively). When Na+ was reintroduced at the time of thapsigargin removal, a significant change in the slope was observed (P < 0.001). In contrast, no significant change in the slope was observed when the absence of extracellular Na+ was maintained at the time of thapsigargin removal (P > 0.4). As compared to that of thapsigargin, the effect of caffeine on [Ca2+li was more heterogeneous. It was also characterized by the frequent occurrence of
of Ca” from intracellular stores [15]. By contrast, in the presence of Ca*+ and the absence of Nao+(Fig. 5, right panel), the increase in “%a outflow induced by caffeine was almost completely suppressed. The figure shows that reintroduction of Na+ into the extracellular milieu induced a marked increase in *%Zaoutflow from islets exposed to caffeine as in the case of thapsigargin. These data indicate that in the presence of extracelluar Na’, the Ca2+ released by caffeine from in~acelIular stores was extruded from the ceil presumably by Na/Ca exchange. When such a process was blocked by the absence of extracellular Na+, higher [Cs2”]i levels were reached in response to caffeine as in the case of thapsigargin. When the process was reactivated by Nao+ readmission, 45Ca outflow was stimulated (Fig. 5 right panel) and [Ca*‘]i dropped (Fig. 4). The removal of extracellular Na+ may not solely inhibit Na/Ca exchange but could also interfere with
11
Na/Ca exchange in the pancreatic B cell
other cellular processes, e.g. Na/H exchange, Na+,K+-ATPase, membrane potential. . . However, alterations in the latter processes do not appear to be responsible for the observed changes in [Ca*+]i and 45Ca outflow Indeed, at low glucose concentrations (~2.8 mM) , extracellular Na+ removal only barely affects membrane potential [16]. N& removal could inhibit Na+,K+-ATPase, but this would not lead to Na+i accumulation and hence would not affect cellular Ca2+ movements [7, 171. Na$ removal could also inhibit Na/H exchange and result in a decrease in pHi [18]. This acidification may reduce Ca2+ outflow but to a much lower extent than Nat removal [19]. In addition, the decrease in pHi reduces 45Ca outflow by inhibiting Na/Ca exchange [ 191. Our data provide direct evidence that Na/Ca exchange, working in its Ca2+ efflux mode, may regulate the [Ca2+]i of the pancreatic B cell. Thus, the presence of extracellular Na+ favoured the decrease in [Ca*+]i that had been raised by two different mechanisms. Our data strongly indicate that the presence of Nao+probably acted by increasing Ca2+ outflow from the cell. This conclusion is strengthened by the observation that the absence of Na$ affected the late instead of the immediate effects of thapsigargin on [Ca2+]i. In other words, the absence of NaJ did not affect the action of the drug (blockade of Ca2+ uptake by the endoplasmic reticulum). Rather, it interfered with a cellular homoeostatic mechanism allowing the B cell to recover from a cytosolic Ca*+ load. Our data are in agreement with recent patchclamp experiments combined with microfluorimetric recordings in mouse pancreatic B cells showing that Na+ removal produced both an elevation of resting [Ca2+]i and an increased amplitude of depolarization-evoked [Ca2+]i transients [20]. In conclusion, the B cell Na/Ca exchange appears as a Ca pumping system designed to extrude Ca*+ when its intracellular concentration is raised above basal levels, as in many other types of cell. Further work is required to determine the exact kinetic constants of the exchange and the possible participation of the process in Ca2+ uptake mechanisms. Acknowledgements-The
authors are indebted to Ch. Pastiels, M. Hermann and A. Van Praet for technical help, and to P. Surardt for secretarial help. REFERENCES 1. Dipolo Rand Beau& L, Ca2+ transport in nerve fibers. Biochim Biophys Acta 947: 549-569, 1988. 2. Reeves JP and Philipson KD, Sodium-calciumexchange activity in plasma membranes vesicles. In: Sodiumcalcium Exchange (Eds. Allen TJA, Noble D and
Reuter H), pp. 27-53. Oxford Science Publications, Oxford, 1989
3. Sheu SS and Blaustein MP, Sodium/calcium exchange and regulation of cell calcium and contractility in cardiac muscle, with a note about vascular smooth muscle. In: The Heart and the Cardiovascular System (Ed. Fozzard HA), pp. 50!&535. Raven Press, New York, 1986. 4. Eisner DA and Lederer WJ, Na-Ca exchange: stoichiometry and electrogenicity. Am J Physiol 248: C189-C202, 1985. 5. Mullins LJ, The generation
of electric currents in cardiac fibers by Na-Ca exchange. Am J Physiol236: c103-CllO, 1979. 6. Hellman B, Andersson T, Berggren PO and Rorsman P. Calcium and nancreatic &cell function XI. Modification of “Ca’fluxes by Na removal. Biochem Med 24: 143-152, 1980. 7. Herchuelz A, Sener A and Malaisse WJ, Regulation
of calcium fluxes in rat pancreatic islets: calcium extrusion by sodium-calcium countertransport. J Membr Biol57: 1-12, 1980. 8. Janjic D and Wollheim CB, Interactions of Ca*+, M$+,
and Na+ in regulation of insulin release from rat islets. E222-E229, 1983. 9. Siegel EG, Wollheim CB, Renold AE and Sharp GWG, Evidence for the involvement of Na-Ca exchange in glucose-induced insulin release from rat pancreatic islets. J Clin Invest 66: 9961003, 1980. 10. Gobbe P and Herchuelz A, Does glucose decrease cytosolic free calcium in normal pancreatic islet cells? Am J Physiol244:
Res Commun 1989.
Chem Path01 Pharmacol
63: 231-247,
11. Plasman PO, Lebrun P and Herchuelz A, Characterization of the process of sodium-calcium exchange in nancreatic islet cells. Am J Phvsiol259: E844-E850. , 1990. 12. Herchuelz A, Pochet R, Pastiels C and Van Praet A, Heterogeneous changes in [Ca2+]iinduced by glucose, tolbutamide and K+ in single rat pancreatic B cells. Cell Calcium 12: 577-586, 1991. 13. Carafoh E, Membrane transport of calcium: an overview. Methodr Enzymol157: 2-11, 1988. 14. Thastrup 0, Cullen PJ, Drobak PK, Hartley MR and
Dawson AP, Thapsigargin, a tumor promoter, discharges intracellular Ca2’ stores by specific inhibition of the endoplasmic reticulum Ca2+-ATPase. Proc Nat1 Acad Sci USA 87: 2466-2470,
1990.
15. Endo M, Calcium release from sarcoplasmic reticulum. Cm-r Top Membr Transp 25: 181-230, 1985. 16. Ribalet B and Beigelman PM, Effects of sodium on /Jcell electrical activity. Am J Physiol242: C296C303, 1982.
17. Lebrun P, Malaisse WJ and Herchuelz A, Na+-K+ pump activity and the glucose-stimulated Ca2+-sensitive K+ permeability in the pancreatic B-cell. J Membr Biol 74: 67-73, 1983.
18. Lebrun P, Plasman PO and Herchuelz A, Effect of extracellular sodium removal upon %Rb outflow from pancreatic islet cells. Biochim Biophys Acta 1011: 611, 1989. 19. Lebrun P, Malaisse WJ and Herchuelz A, Effect of the absence of bicarbonate won intracellular oH and calcium fluxes in pancreatic Islet cells. Biochim ‘Biophys Acta 721: 357-365,
1982.
20. Rorsman P, Ammall C, Berggren P-O, Bokvist K and Larsson 0, Cytoplasmic calcium transients due to single action potentials and voltage-clamp depolarizations in mouse pancreatic B-cells. Embo J 11: 2877-2884,1992.